90 GIST THE EFFECT OF THE INTERNAL FRICTION OF FLUIDS
59. On account of the difficulty which Baily experienced in obtaining accurate results with the long rods arid spheres attached, he divided the brass and steel rods near the centre of oscillation, and after having cut off an inch from each portion inserted tbe spheres where the rods had been divided. The results thus obtained constitute the loth set of experiments. He afterwards removed the lower segments of the rods, and obtained the results contained in the 16th set. I shall give the computation of the latter set first, inasmuch as the formulae to be employed are exactly the same as those required for the 14th set. The experiments belonging to this set in which the spheres were swung with iron wires have already been computed under the head of spheres attached to fine wires.
Baily's results with the spheres at the end of the short rods.
No. 60 —1^-inch sphere with brass rod. No. 61 — 2-inch sphere with ditto. No. 62 —3-inch sphere with ditto. No. 63 - 1^-inch sphere with steel rod. No. 64 - 2-inch sphere with ditto. No. 65 -3-inch sphere with ditto.
Value of n Weiftht of adhesive air
No. By theory By experiment Difference By theory By experiment Difference
60 2-149 2-198 + 0-049, or +7134 1-011 1-047 +0-036, or +JH
61 1-879 1-901 +0-022, or +g^ 1-619 1-513 -0-106, or -^
62 1-787 1-830 +0-043, or +-/0- 3-970 4-202 +0-232, or +^-
63 1-960 1-904 -0-056, or - ,jV 0-570 0-537 -0-033, or --A-
64 1-796 1-785 -0-011, or -r^ 1-239 1-227 -0-012, or -r<h
65 1-758 1-779 + 0-021, or +y\- 3-609 3-720 +0-111, or +1V
Here again the differences between theory and experiment are extremely small. In the case of the pendulum No. 61, Baily's two results 1*901 and 1*513 appear to be inconsistent, as riot agreeing with the formula (155).
60. The following table contains the values of T, k, and &, deduced from the given data, and employed in the calculations of which the results are contained in the two preceding tables. It is added, partly to facilitate a comparison of the circumstances of the different experiments, partly to assist in the re-computation